good health at low cost revisited - ghlc...benjamin palafox lshtm chapter 8 good health at low cost...

GOOD HEALTH AT LOW COST 235

KeralaCosta Rica

Sri Lanka

China

Benjamin Palafox

LSHTM

Chapter 8

GOOD HEALTH AT LOW COST REVISITEDFurther insights from China, Costa Rica,

Kerala and Sri Lanka 25 years later

■ Introduction

In Chapter 1, we saw how China, Costa Rica, Kerala and Sri Lanka, the fourlow-income case studies profiled in the original Good health at low cost reportfrom 1985, had achieved remarkable health gains by the early 1980s. Althoughthey attained these gains in different ways, there were some important similari-ties that offered crucial evidence in support of the principles advocated at theAlma-Ata conference in 1978 and provided insight into ways that might reduceinfant, child and maternal mortality.

From the health system perspective, key factors that emerged were long-term(and above average) investment in financial and human resources for health, espe-cially in primary care; strong political commitment to good health for the wholepopulation; a high degree of community involvement; and equity of access and use.In addition, each country had enacted policies beyond the health system, imple-menting wide-ranging policies which addressed many different determinants ofhealth, with a particular emphasis on expansion of education, especially forgirls1.

The case studies also demonstrated the value of integrating services both hori-zontally and vertically, ensuring the inclusion of prevention within essentialprimary care and the necessary linkages between primary care and the rest of thehealth system. In Kerala and Sri Lanka, the expansion of essential primary healthservices, with a focus on maternal and child health, was considered critical forthe reduction of both child and maternal mortality. Results were attributed to an emphasis on integrated service provision models that improved access to antenatal and postnatal care, and dramatically increased rates of institutionaldelivery and use of skilled birth attendants. The care of young children wasboosted by measures to improve rates of immunization and effective manage-ment of communicable disease. Costa Rica’s early reforms strengthened primarycare (with a focus on family planning) and improved access to higher levels ofcare, extending coverage of immunization, and improving nutrition and sanita-tion. China also saw improvements in maternal and child health. As well asimplementing maternal and child health interventions like those in the othercountries, it engaged in a series of nationwide campaigns to control the vectorsof communicable diseases, improve sanitation and increase access to clean water. A key feature of the Chinese approach was the involvement of barefootdoctors, a cadre of village health workers working to improve health within theircommunities.

By the early 1980s, after years of implementing these policies, China, CostaRica, Kerala and Sri Lanka had achieved life expectancy approaching that ofsome high-income countries. However, with this came an epidemiological

GOOD HEALTH AT LOW COST REVISITED

GOOD HEALTH AT LOW COST236

transition that rapidly increased the burden of chronic and other noncommuni-cable conditions, introducing a new set of challenges for health policy-makers.Adapting health systems to these new realities has been complicated by a numberof important contextual changes since the mid-1980s, including uneveneconomic growth, political and economic crises, changing international tradeflows, the emergence of new technology, and migration, all of which conspiredto widen inequalities in wealth and, consequently, health. With relatively low percapita expenditure on health, how could they sustain progress already made,further reduce inequities in health and cope with the higher health care costsassociated with an ageing population and changing lifestyles? Can these coun-tries still be regarded as models of population health improvement by otherdeveloping countries?

In this chapter, we revisit the four original case studies and ask how each countryhas fared since 1985. We review the progress countries have made in improvinginfant and maternal mortality (as key indicators of health system performance),describe the main changes to their health systems and broad sociopoliticalcontexts, and examine the possible mechanisms through which these changesmay have influenced population health. (Box 8.1 outlines the search strategy anddata sources used to conduct this desk review, and Chapter 2 gives additionaldetails on the conceptual framework, research approach and analyticalmethods.) We conclude by seeking lessons that can be learned from their experiences.

■ China

In 1978, Deng Xiaoping initiated wide-ranging economic reforms that sweptaway many elements of the centrally planned Chinese economy established byMao Zedong. The introduction of free markets ushered in a process of rapidtransition that set the country on track to become an economic powerhouse. Inless than a decade, the country’s economic base shifted away from agriculture toindustrial production, with much of its output sold abroad. Since 1985, it hasmaintained a remarkable rate of economic growth of nearly 10% a year2. Thisachievement reflected many factors, but one of the most important was theestablishment of special economic zones that allowed foreign investors to takeadvantage of low labour costs and favourable tax regimes. This facilitated theexplosion of industrial manufacturing and fuelled the labour market in thesezones and in urban areas in general. As a result, migration from rural to urbanareas has increased the size of China’s cities to meet the new labour demands,although more than half of its 1.3 billion people still live in rural areas.



China’s ascendancy in the global economy has been matched by progress inother areas. Since the late 1970s, it has managed to slow population growth tobetween 0.5% and 1.6% per year, and by 2008, the total fertility rate had fallento 1.78 births per woman2. This is partially the result of the one-child policyintroduced at the same time as the economic reforms. China has improved otherkey development indicators, particularly literacy, poverty and basic education.These developments were mirrored in health gains: between 1985 and 2008,overall life expectancy rose from 66.9 to 73.1 years, while vaccination coverageagainst diphtheria, pertussis and tetanus (DPT) among children under two yearsof age increased from 78% to 97% (Table 8.1).

The period from the 1950s to the early 1980s has been regarded as China’s water-shed period for health, when enormous gains were achieved. Life expectancy



Box 8.1 Desk review search strategy and data sources

We conducted a review of published, peer-reviewed literature using the PubMed and

EconLit bibliographic databases. Relevant sources were searched first using both

standardized terms and keywords based on the outcomes of interest (e.g. infant, child

and maternal mortality). The results were then combined with searches based on

relevant determinants of health. Health systems determinants were informed by the

WHO Health Systems building blocks of service delivery, health workforce, information,

medical products and technologies, financing, leadership. Non-health systems determi-

nants were related to public expenditure, economic policy, rule of law, water and

sanitation, education policy, social security, gender policy, public administration for

public provisioning; plus structural factors (e.g. system of government, media, food

supply, etc.), situational factors (e.g. elections, conflict, natural disasters, migration,

etc.), cultural factors (e.g. religious values, accepted forms of hierarchy, awareness of

rights, trust in institutions, etc.) and international or exogenous factors (e.g. foreign aid,

international trade agreements, influence of civil society organizations, etc.) likely to

influence policy. The results were then combined and limited to those pertaining to the

four case countries for the years 1985–2009. Titles and abstracts of the remaining

sources were screened for relevance and a review of bibliographies was conducted

among selected documents to identify additional sources. This was supplemented with

key informant interviews of national and international experts familiar with the case

study contexts, and with a search of relevant grey literature since 1985. For this, we

used similar keywords to search various document repositories such as the Eldis and

British Library for Development Studies websites, and also those of multilateral

organizations, such as WHO and the World Bank.

increased across the entire country. After this period, China continued to experi-ence health gains in some areas, most notably in maternal mortality, where adramatic fourfold reduction has been achieved since 1980, reaching 40 per100000 live births in 20083. However, in other areas, particularly child mortal-ity, China has not performed as well4. Following China’s impressive gains ininfant mortality described in the original Good health at low cost, progress haltedfor nearly a decade, remaining at approximately 40 deaths per 1000 live births,until resuming a downward trend during the late 1990s. By 2008, this indicatorhad stabilized at a rate of approximately 15 deaths per 10005,6 (Figure 8.1).

The slower progress in infant mortality has been attributed to growing healthinequalities, linked to the sweeping changes in the economy. There were winnersand losers, with the winners concentrated in the areas undergoing the greatest



Table 8.1 Selected development indicators, China vs. middle-income countries

Indicator China Middle-income countriesa

1985 2008 1985 2008

Vaccination, DPT (% of children aged 12–23 months)

78 97b 47 81

Primary school completion rate, total(% of relevant age group)

n/a 96.0 n/a 92.3

Poverty gap at $2 a day (purchasing power parity) (%)

47.3c 12.2d n/a 10.7e

Literacy rate, adult total (% of people aged 15 and above)

65.5f 93.7 n/a 82.7

Fertility rate, total (births per woman)

2.64 1.78 3.68 2.43

Life expectancy at birth, total (years)

66.9 73.1 62.7 68.5

Source: Data from reference 2.

Notes: n/a: Not available; DPT: Three doses, diphtheria, pertussis and tetanus; aBased on WorldBank income grouping; bValue is for 2009; cValue is for 1984; dValue is for 2005; eValue is forupper-middle-income countries, as defined by the World Bank (for comparison, the value forlower-middle-income countries is 54.1%); fValue is for 1982.

economic reform7,8. A 2009 review of health and health care since economicliberalization found deepening inequalities between urban and rural areas, andamong income groups9. For example, Shanghai, China’s leading commercialcentre, saw an improvement in life expectancy of four years between 1981 and2000, to 78 years; while in Gansu, one of China’s poorest provinces, theimprovement was of only 1.4 years over the same period. Consequently, by2000, a 13-year gap in life expectancy had opened up between the two regions;and when plotted against provincial GDP, a clear gradient in life expectancy wasapparent10. Similar patterns were observed with infant mortality rates in ruralareas. Rates were nearly five times higher in the poorest counties than in thewealthiest ones. These were also mirrored in under-5 mortality. Between 1996and 2004, a sixfold difference emerged between the highest and lowest socio-economic quintiles, with a fall of 50% among wealthy rural populationscompared with only 16% among the least wealthy groups10.

The original Good health at low cost report linked China’s remarkable healthgains with its relatively well-developed social welfare system. In rural areas, wheremost Chinese people lived at the time, the commune played a central role. Itowned the land and managed its use. It also administered the Rural CooperativeMedical Care System (RCMCS), a system that provided members of thecommunity with a basic form of health protection. Basic curative, preventive



Figure 8.1 Infant and maternal mortality, China, 1970–2009

Sources: Data from references 3, 5 and 6.

0

20

40

60

80

100

120

140

160

18019

70

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

Maternal mortality ratio (per 100 000 live births)

Infant mortality rate (per 1000 live births)

and public health services were largely delivered through health centres ownedby the commune and operated by barefoot doctors11. By the 1970s, 90% of thepopulation had health coverage, either from the RCMCS in the rural areas orfrom different state-owned enterprises in the cities (Figure 8.2).

Health status in rural areas continued to improve immediately following theprivatization of agricultural production; this has been attributed to improvementsin agricultural productivity that increased not only household income in rural



Figure 8.2 Timeline of key events influencing health, China, 1975–2010

Nearly 90% of populationcovered by either health insuranceor under Rural CooperativeMedical Care System

Launch of Basic MedicalInsurance scheme for all urbanformal sector employees

China becomes member ofWorld Trade Organisation

Introduction of New RuralCooperative Medical Care System

One-child policy

Introduction of economicliberalization, devolution and privatization of health system begins; decollectivization of communes and dismantling of public health system

Introduction of mandatory medical savings accounts for employees of private and state-run enterprises

Epicentre of SARS pandemic in Guangdong, China

Introduction of new urban insurance schemes

1975

1980

1985

1990

1995

2000

2005

2010

areas but also nutrition7,8. However, these gains were short lived as the spread ofeconomic liberalization left much of the rural population uninsured. Althoughmany of those involved initially welcomed privatization of agricultural landpreviously owned by the communes, the process destroyed the economic basisupon which the RCMCS had operated12. Simultaneously, the central govern-ment reduced its investment in health care and other public services. Between1978 and 1999, its share of national health care spending fell from 32% to15%12. The areas it withdrew from were taken over by provincial and localauthorities, who were required to fund them from local taxation. This favouredwealthy coastal provinces that had stronger tax bases over less wealthy ruralprovinces, and laid the basis for major and growing disparities between invest-ment in urban and rural health care12.

These overall reductions in funding had many negative consequences for thequality and affordability of local public health services. Out-of-pocket expendi-ture began to rise as health facilities relied increasingly on the sale of services togenerate sufficient operating revenue, which was exacerbated by ill-conceivedincentives, such as a salary bonus scheme that linked the size of the bonus tooverall facility revenues12,13. Over the period of reform, income and the relativecost of treatment became increasingly important predictors of infant mortalityas the health system began to rely more heavily on private expenditures14. By2003, private expenditure reached 63% of total health spending, and 92% ofprivate spending was out of pocket; however, by 2007, the proportion of spend-ing from private contributions had declined to 55%2.

Much of this decline has been the result of increasing government investment inthe health sector made possible by the massive economic growth that has boostedgovernment revenues. China’s ageing population, the increasing prevalence ofcatastrophic health care costs, the severe acute respiratory syndrome (SARS)pandemic that originated in southern China, and the rising demand for ruralhealth services have since made health a top government priority. One of themost significant developments has been the New Rural Cooperative MedicalCare System (NRCMCS), introduced in 2005. While similar in spirit to itspredecessor, which had become defunct by the late 1970s, this new programmeis voluntary and the pooled risk fund is fed by members’ contributions and bysubsidies from central and local government. It works at the county level (muchlarger than the old communes) and focuses on protecting members from cata-strophic medical expenses related to inpatient care (where the original RCMCSprovided basic curative and preventive services).

In urban areas, formally employed residents have benefited from schemes imple-mented since 1995, such as the Basic Medical Insurance package and mandatory



medical savings accounts. However, these benefits excluded dependents, andlarge groups of urban residents, particularly economic migrants from rural areas,were left mostly unprotected. To address this gap, the government began to scaleup its Urban Resident Basic Medical Insurance scheme in 2008. The voluntaryprogramme enrolls entire households, to target children, the elderly, the disabledand other non-working urban residents. Like the NRCMCS, the scheme isfunded by contributions and also by premium subsidies from the government.

The government has also attempted to reduce the cost of care by encouragingthe use of lower-level facilities, although perceptions of poor-quality service atthese lower levels still remain a barrier. In response, government funds are beinginvested in new primary, preventive and rehabilitative centres and in renovatingolder village clinics and township health centres. Funds are also being used fortraining, advertising health facilities and improving community participation inthe health system. The NRCMCS includes representatives of the farmers andvillage committees served by the programme and the new scheme is undercounty-level management, making it more accountable and closer to those whoaccess the benefits of the programme (Bloom G, personal communication,2010).

At this early stage, definitive evidence of the impact of these reforms on healthand service utilization is not yet available. A review of some recent studies indi-cates that adverse selection may be a problem with the new insurance schemes,skewing enrolment towards those already unwell9. Another review of pilotstudies of these insurance schemes has shown only moderate protection fromcatastrophic spending and limited protection for the poorest beneficiaries, asout-of-pocket spending remains a problem15. This has not, however, discour-aged the Chinese Government, which has targeted universal health insurancecoverage as a priority12. With nearly 90% of rural residents covered by theNRCMCS (accounting for 815 million people), and 65% of urban residentscovered by the corresponding urban scheme by 200816, this target appears to bewell within reach.

■ Costa Rica

Costa Rica has long been recognized as one of the most politically and econom-ically stable countries in Latin America. Since 1985, there has been steadyannual growth in GDP, often as high as 8 to 9%2,17. Despite having a per capitaGDP that is merely average for an upper-middle-income country (US$6564 in2008)2,17, this small nation of fewer than five million people has consistentlybeen among the top Latin American countries in terms of the Human



Development Index (HDI), a multidimensional measure of social and economicdevelopment that combines indicators of life expectancy, educational attainmentand income: ranking 62 in the world and sixth among Latin American countriesin 201018. While many development indicators were already quite good in1985, most have continued to improve since then and have even surpassed othercountries in the region with comparable income levels, such as Panama (shownwith Costa Rica in Table 8.2). For example, by 2008, the adult literacy rate was96%, less than 2% of the population was living below the international povertythreshold of US$2 (adjusted for purchasing power parity) per day, more than95% of the population had access to improved water and sanitation, and thetotal fertility rate was 1.96 births per woman2.

Costa Rica has also maintained its impressive performance with respect to thehealth indicators documented in the original Good health at low cost, surpassing



Table 8.2 Selected development indicators, Costa Rica vs. Panama

Indicator Costa Rica Panamaa

1985 2008 1985 2008


90 86b 73 82


77.3 92.9 82.4 n/a

Poverty gap at US$2 a day (purchasing power parity) (%)

8.6c 1.3d 13.1e 7.06f


92g 96 88h 94


3.46 1.96 3.34 2.55


74.6 78.9 71.4 75.7


Notes: n/a: Not available; DPT: Three doses, diphtheria, pertussis and tetanus; aSelected as suitablecomparator due to similar location, population, total GDP and GDP per capita; bValue is for 2009;cValue is for 1986; dValue is for 2007; eValue is for 1991; fValue is for 2006; gValue is for 1984;hValue is for 1980.

all other countries of the same income level within the region. In 2008, lifeexpectancy of 81.4 years for women and 76.6 years for men was second only toCanada in the western hemisphere2; the probability of maternal death was estimated to be approximately 25 per 100000 live births; and infant mortalityhad declined steadily and is now estimated to be approximately 9.6 per 1000 live births, representing a sevenfold reduction over a three-decade span3,5,6

(Figure 8.3).

This continued improvement is attributed to Costa Rica’s long history of invest-ment in social welfare, perhaps best represented by the Costa Rican SocialSecurity Fund (CCSS), which was one of the first publicly administered socialinsurance models introduced in the region during the early 1940s. In additionto administering the national pension and other social security programmes, thisautonomous public body also delivers most medical services free at the point ofdelivery, providing a comprehensive package of medical insurance benefits. Thebulk of primary care is delivered through health centres and clinics that provideoutpatient services, family and community medical services, and health promo-tion and prevention programmes, referring patients to higher levels of care asrequired. While the private sector is small in Costa Rica, public facilities mayrefer patients to the private sector when they are overloaded, or patients maychoose to see a private physician to avoid long waiting times17. To address some



0

10

20

30

40

50

60

70

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1996

1994

1998

2000

2002

2004

2006

2008

2010



Figure 8.3 Infant and maternal mortality, Costa Rica, 1970–2009


issues of insufficient capacity in the public sector, the CCSS also contracts outsome services to private entities, mainly to health cooperatives for primary serv-ices in urban areas, but also to private laboratories for diagnostics. In addition tooverseeing and regulating the health system, the Ministry of Health sharesresponsibility for public health service delivery with the CCSS.

Total health expenditure as a proportion of GDP has remained consistent atapproximately 8% (slightly higher than the average of approximately 7% fordeveloping countries in Latin America and the Caribbean)2. In 2007, 27% ofthis expenditure was private, 85% of which covered out-of-pocket payments forambulatory care in the private sector (Rosero-Bixby L, personal communication,2010)19. Two thirds of the 73% public health expenditure was from the CCSS,making it the country’s most important source of health financing2,19. As anindependent public institution, the CCSS is financed primarily by contributionsfrom employers (9.25% of payroll) and workers (5.5% of wages). Followingworker protection legislation introduced in 2000, the self-employed are requiredto contribute 4.75% of their reported income, and the poor are covered byseveral subsidized schemes19. By 2006, 88% of the population was covered bythe CCSS and 93% of the population had adequate access to primary care services19.

Health system reforms since 1985

In addition to the factors already noted in the original Good health at low costvolume, including sustained public health expenditures, political stability andcommitment, clear national consensus on the role of the health system andpopular support for the CCSS17,20–22, several reforms that were implementedfrom 1994 onwards further strengthened the Costa Rican health system and havebeen linked with improved health outcomes (Figure 8.4). These reforms, whichfollowed the vision for the health system set out in the 1970s, have been associ-ated with reductions of 8% and 2% in child and adult mortality rates, respec-tively20, and fall into two main categories: further extending coverage and qualityof primary care, focusing on underserved areas; and further improving themanagement, financing and delivery of medical services under the CCSS.

Achieving universal access to primary care, particularly in underserved ruralareas, was greatly facilitated by the introduction of the EBAIS community clinics(Equipos Básicos de Atención Integral en Salud). Each clinic is responsible for ageographical area that covers approximately 4000 people and offers a full rangeof primary care, health promotion and preventive services. Where necessary, the EBAIS is mobile. At a minimum, EBAIS clinics are staffed by a doctor, anurse and a technician, who are supported by personnel from the higher-level



administrative grouping (known as the health area) to which the EBAIS belongs.These can include laboratory technicians, social workers, dentists, nutritionists,pharmacists and medical records specialists23. During the first stage of imple-mentation, 232 EBAIS were established in 1995, with priority given to the mostunderserved communities. By 2004, there were a total of 855 EBAIS across thecountry19.



Figure 8.4 Timeline of key events influencing health, Costa Rica, 1985–2010

Further split of responsibilitiesbetween Ministry of Health andCCSS

Performance-based contractsintroduced between the CCSSpurchasing directorate and healthfacilities

Act of Worker Protectionrequiring all independent workersto join social health insuranceprogramme

Reforms introducing the EBAIS greatly improving rural access to primary health care

Decentralization of central authority and responsibility to public providers

Deregulation of insurance market

1985

1990

1995

2000

2005

2010

The literature is characterized by broad consensus on the favourable impact thatexpansion of primary care has had on improving equitable access in Costa Rica.Prior to 1994, access to primary care was restricted to approximately 25% of thepopulation. One analysis found that in the areas where reforms were imple-mented after 1995, the percentage of the population with adequate access tohealth services had risen from 64% to 79% by 200020, while the national rateof health coverage rose to 69% over the same period. Today, coverage is nearlyuniversal17,19,23,24. Data presented in the original Good health at low cost showedthat, between 1972 and 1980, 41% of the decline in infant mortality could beattributed to primary care interventions, while an additional 32% was due toimprovements in secondary care. Socioeconomic progress and declining fertilityrates explained the remaining decline25. In 1991, further analysis supportedthese original conclusions26. Although there is little evidence directly linkingcontinued health gains in Costa Rica since 1985 to continued improvements inprimary care access, the country’s previous experience suggests that it may becontinuing to play a part.

While the EBAIS greatly enhanced the physical reach of the CCSS, other impor-tant reforms implemented since the 1990s have focused on the organization’sadministrative structure. One such reform was the creation of a purchasing divi-sion within the CCSS, further separating the financing, purchasing and serviceprovision functions of the organization. This allowed for improvements inquality and efficiency, such as the shifting away from a historical budgetingapproach to resource allocation towards one intended to enhance production,user satisfaction and clinical practice, based on performance managementcontracts between the newly created purchasing division and serviceproviders17,27.

Another important reform during this period was the 1998 Law onDecentralization28, which aimed to improve health system responsiveness bymeans of administrative decentralization of the CCSS. One of the mechanismssupporting this transfer of power was the creation of democratically electedcommunity health boards to supervise the delivery of local services. This broad-ened community participation as local decision-makers became involved insetting priorities and performance targets for health17,19.

However, the financial sustainability and equity of Costa Rica’s state-drivenmodel remain pressing issues as the cost of financing the health system contin-ues to increase with the ageing population and the changing burden of disease.Despite the large operating revenues provided through member contributions,the government continues to commit substantial portions of its annual budgetto health. For example, more than a fifth of total government expenditure went



to health between 2003 and 20072. As the success of the health system isdependent on the principle of solidarity to maintain high participation rates inthe scheme across all population groups and enhance the progressive characterof the overall system, low rates of affiliation (52%) among the economicallyactive population and the high prevalence of contribution evasion amongemployers and workers alike threaten financial sustainability19. Until recently,the government of Costa Rica resisted the introduction of private insurance,largely because it did not believe it was in the citizens’ best interest and becauseit feared that, once introduced, foreign corporations would quickly dominate theentire insurance market27. From this perspective, the deregulation of the healthinsurance market in 2009 to allow private medical insurance creates an obviousrisk to equity, with scope for wealthy and healthy patients to opt out of publiclyfunded care, so undermining popular support for the CCSS. Costa Rica’s expe-rience will undoubtedly continue to yield further insight into best practices inhealth financing.

■ Kerala

For most of the 20th century, the economy of this southern Indian state laggedbehind much of the rest of the country. Between 1970 and 1987, Kerala’s annualgrowth in net domestic product was an average of 1.9%, nearly half of the all-India figure. But post-1987, the state’s economy grew at a rate of 5.8% per year,and by 2000, its per capita income was 20% higher than the all-India figure 29.This growth has largely been driven by the service sector, related to transporta-tion, trade, hotels, restaurants and telecommunications, rather than the moreconventional production of commodities. The increasing demand for these serv-ices has been linked to the increase in disposable income and ownership of assetssuch as homes, vehicles and appliances, which were supported largely by thehuge influx of remittances from Keralites who since the 1970s have migrated towork in other parts of India and in the Gulf States30.

Underlying Kerala’s remarkable change in economic growth was the state govern-ment’s long political commitment to investment in social welfare and equality, ascharacterized by the development of universal access to education, strong labourorganization and popular movements promoting dialogue among castes. Since itsformation in 1956, Kerala has consistently ranked higher on the HDI than allother states in India. By 2005, Kerala had nearly achieved universal elementaryeducation and had attained a gender ratio of 1.058 females to every male: iden-tical to that in Europe and North America, but quite different from many otherparts of India where selective female abortion is widespread29. Gaps in humandevelopment also continue to close across gender and social groups. Census data



from 1961 to 1991 show that the literacy gap between the general and ruralscheduled caste populations (representing those on Kerala’s social margins) hasconsistently narrowed. The same data also show that the state’s growth in literacywas higher than in all other Indian states, with the greatest relative gains amongwomen. The availability of schools and good road networks, typical outputs ofthe state government’s past investments, were identified as key factors in explain-ing the observed gains in literacy31 (Table 8.3).

This continued performance on development indicators is also mirrored inKerala’s improving population health, better than all other Indian states. In1980, overall life expectancy at birth was 66 years35, and by 1995, it had risento 70.4 years for males and 75.9 years for females compared with Punjab, whichduring the same period had the next-best life expectancy across India, at 66.7years for males and 68.8 years for females29. Infant mortality experienced adramatic decline and more than halved from 1981 to 2005–2006, when it wasestimated at approximately 15 deaths per 1000 live births (Table 8.4). By



Table 8.3 Selected development indicators, Kerala vs. all India

Indicator Kerala India

1992–1993 2005–2006 1992–1993 2005–2006


n/a 84.0 46.9 55.3

School attendance (% of children 6–10 years)

94.8 98.4 68.4 82.9

Literate persons (% of total population aged 15–49)

89.8a 93.5b 52.2a 63.4b

Undernutrition prevalence, weight forage (% of children under 5 years)

28.5c 22.9 53.4c 42.5


2.0 1.9 3.4 2.7

Sources: Data for 1992–1993 from reference 32, unless stated otherwise; Data for 2005–2006from references 33 and 34.

Notes: n/a: Not available; DPT: Three doses, diphtheria, pertussis and tetanus; aValues are for 1991and data from reference 29; bValues are derived from combining weighted national estimates forwomen and men; cValues are for children under 4 years.

comparison, Maharashtra, the state with the second-lowest infant mortality rate,experienced 48 per 1000 live births in 200029. It is also remarkable that infantmortality rates in Kerala show almost no difference between rural and urbanareas, unlike the rest of the country where a large gap persists29. While reliabletime series data do not exist for maternal mortality in Kerala, in 2006, it was esti-mated to be 95 deaths per 100000 live births, approximately one third of theestimated rate for India as a whole38.

Despite its low mortality overall, Kerala now has some of the highest rates ofnoncommunicable disease mortality and morbidity in the country39,40. A recentstudy of adult mortality patterns within a rural community showed that coro-nary heart disease has now overtaken communicable diseases to become theleading cause of death in the state, and that the burden of coronary heart diseasedeaths now exceeds that of industrialized countries41. The prevalence of obesityis also rapidly increasing, and Kerala has the second highest rate of obesityamong women of all states in India (21% with body mass index of 25+, whilethe national average is 11%)29. Trends in alcohol consumption are also a causefor concern because, although the overall Indian average is low, consumption inKerala is the highest in the country, at more than double the all-Indiaaverage29,42.

Health system changes since 1985

Kerala’s current health system is composed of parallel public and private sectors.While traditional medicine is important in the state’s health system, the share ofmodern (that is, allopathic or western) health services is highest in Kerala amongall Indian states43. The public sector has a well-developed network of healthfacilities (a legacy of Kerala’s prior investment in social welfare), with nearly 200hospitals and more than 1000 primary health facilities, each staffed with a



Table 8.4 Infant and maternal mortality, Kerala, 1981 to 2005–2006

Health indicator 198136 1992–332 1998–937 2005–633


39.1 23.8 16.3 15.3

Maternal mortality ratio(per 100000 live births)

n/a n/a n/a 9536

Sources: References 32, 33, 36, 37.

Note: n/a: Not available.

doctor providing a full range of treatment and prevention services (for example,vaccinations and family planning)44. As elsewhere in India, private sector growthincreased dramatically in the early 1980s and quickly surpassed that in thepublic sector (Figure 8.5). For example, between 1986 and 1996, the number ofprivate sector beds rose by 40%, from 49000 to 67500, while the number ofbeds in public facilities grew by only 5.5% over the same period, from 36 000 to3800045. But despite varying degrees of service quality46, a lack of regulation47

and concerns of supplier-induced demand48, the private sector now handlesmost of the caseload in the state and has also surpassed the public sector in otherareas, including the availability of advanced diagnostics, such as magnetic reso-nance imaging. However, unlike Sri Lanka, where private-sector outpatient services tend to complement the hospital-dominated public sector, the privatesector in Kerala offers a mix of services that are in direct competition with thepublic sector.

The shift from the public to private sector was facilitated by a number of devel-opments. Fiscal crisis in the 1970s and in the 1990s led to the introduction ofpoorly implemented cost-recovery mechanisms (i.e. user fees) that generatedinsufficient operating revenues. The fiscal crisis also decreased health budgets,and funds earmarked for health were increasingly used to meet salaries. Between1985 and 2003, the share of health in the state revenue budget fell from 7.7%to 5.4%49. Shortages of medicines and other consumables decreased the qualityof public-sector services and negatively impacted upon popular confidence inthe government-funded health system, encouraging patients to seek privatehealth care. Increased purchasing power among poorer groups (brought aboutby increasing incomes across all socioeconomic groups and decreasing fertilityrates) further fuelled the already high demand for modern health services, somuch so that by the mid-1980s health service use – both among low- and high-income groups – was already shifting towards the private sector.

These factors have compounded over the years and the effect is clearly seen incurrent patterns of health spending. Compared with all other Indian states,Kerala spends at least twice the annual amount per household, at nearly US$ 38per capita in 2004–2005a; and 86.3% of this falls upon households as out-of-pocket spending. Public funds account for slightly less than 11% of total healthexpenditure, and are raised from both tax and non-tax revenues at national andstate level, with a small proportion also generated from user fees49. Funds fromthe central government are allocated to states to implement national



a Historical exchange rate of US$ 1 to 44.94 Indian rupees was used, averaged over the fiscalyear from 1 April 2004 to 31 March 2005. Exchange rate obtained fromhttp://www.oanda.com/currency/historical-rates.

programmes, while programmes delivered by local governments (panchayats) arefinanced by transfers from the state government.

Not surprisingly, the reliance on household spending has had adverse implica-tions for poor and marginalized groups. One study that looked at householdspending on health across income groups showed that the poor spent 40% oftheir income on health care in 1996, while the rich spent only 2.4%. Compared



Figure 8.5 Timeline of key events influencing health, Kerala, 1955–2010

Formation of the State of Kerala

Effective decentralization ofgovernment health services follow-ing the Panchayati Raj legislation

Establishment of ResourcesCommission to investigate fiscalproblems and suggest solutions

Beginning of 20-year period of‘fiscal crisis’

Growth in government health sector slows, and growth and sizeof private sector surpasses that of public sector

Introduction of user fees in publichealth care

1950

1960

1970

1980

1990

2000

2010

with what they were spending ten years previously, this represents a 450%increase in out-of-pocket expenditure among poor households compared withan increase of 12% for rich households50. A more recent study of inpatienthealth expenditure in rural Kerala showed private spending to be significantlymale biased, as greater amounts tended to be spent on the hospitalization of men(about US$ 129) compared with women (about US$ 93)51. Access in rural areasis further threatened by the persistent underfunding of primary care, and publicsector shortages of doctors and essential commodities, such as drugs, result in therural poor shifting to private health care at a much higher cost52. As in othersettings, health inequities are closely linked to such inequalities in access.

The decentralization reforms introduced with the Panchayati Raj legislation53,54

in the mid-1990s transferred much of the decision-making related to socialwelfare from the central government to the state and local (panchayat) levels.These are considered to be the most likely reforms to address the inequities inhealth that have emerged and have persisted since the 1980s. Theoretically,decentralization of responsibility to the panchayat level was intended to makepublic services, including the health system, more responsive to the communi-ties being served through greater involvement of the community in decision-making processes55. However, the full impact of these reforms on the healthsector has yet to be realized.

One evaluation suggested that decentralization had not yet brought any signifi-cant change to the health sector56. The analysis showed that panchayats in Keralahad, in fact, allocated a lower level of resources to health than what had beenallocated by the state government prior to decentralization. This was largely dueto the absence of sufficient support, innovation and technical expertise at thelocal level to compete effectively for limited panchayat funds. In addition, adirective issued by the State Planning Board barred spending of panchayat fundsfor the purchase of medicines or the maintenance of health facilities, thuscompounding the public-sector quality issues56.

Despite its many years of steady economic growth, greater fiscal pressure mayalso arise as state government revenues continue to be eroded by remittancesfrom foreign workers from which income taxes are not obtained. This is coupledwith the continuing shift away from the primary sector towards the tertiarysector (i.e. the state economy now relies heavily on retail sales, from which it isalso difficult to extract tax revenues) (Acharya A, personal communication,2010). An ongoing challenge for the government will be to ensure that the state’seconomic prosperity is effectively translated into public goods and used toaddress key health issues, such as obesity and chronic disease, and to tackle thewidening inequities that affect access to health services.



■ Sri Lanka

Despite nearly three decades of civil war, Sri Lanka has performed well econom-ically for many years, experiencing steady growth in GDP since 1985, with peakannual growth of 8% in 2006. In addition, with an estimated per capita GDPof US$ 2013 in 2008, this country of 20 million people has the highest percapita income in south Asia2. While its economy historically relied on agricul-tural commodities, over the course of the last century, Sri Lanka has movedsteadily towards an industrialized economy with the development of foodprocessing, textiles, telecommunications and the finance sector. Also, there arenow nearly 1.5 million Sri Lankan citizens working abroad, including many inthe Gulf States and the Middle East. Remittances from these migrant workers,estimated to total US$ 2.9 billion in 2008, are an important source of foreignexchange and have contributed to rising household incomes57.

In the first decade of this century, however, Sri Lanka has had to cope with aseries of challenges. For example, in 2001, Sri Lanka experienced its first-everrecession, a period characterized by power shortages, budgetary problems andintensification of the civil strife that started in the early 1980s. The December2004 tsunami devastated several areas along the southern and eastern coasts ofSri Lanka. A short time later, there was a resurgence of fighting in the ongoingcivil war, continuing until May 2009, when government forces declared theconflict over. Since then, Sri Lanka has experienced a post-war economic boom;however, more than 300 000 people remain internally displaced as a result of theconflict58, and despite being relatively low (5.9% in 2009), unemployment alsopersists, disproportionately affecting women and educated youth57.

Since 1985, Sri Lanka has been able to maintain progress on a number of indi-cators related to human development, largely primed by its early commitmentto social welfare. As a result, a number of these indicators have significantlyimproved, while others, such as poverty and undernutrition, persist. Table 8.5shows key changes, and compares them with India’s.Key indicators on population health outcomes have also generally improvedsince 1985. Total life expectancy in 2008 reached 74 years2; infant mortality hasexperienced a threefold reduction to approximately 10 per 1000 live births, andmaternal mortality is currently estimated to be 30 per 100 000 live births, lessthan half compared with that two decades earlier3,5,6. However, as is clear fromFigure 8.6, the pre-1985 downward trajectory in both of these mortality indica-tors was not maintained. Indeed, at times during the post-1985 period, rates ofinfant and maternal mortality increased before returning to their downwardtrajectory. Furthermore, these health gains have not been equally distributed



throughout the population, and some groups even experienced worsening healthduring this period. Life expectancy for women in 2008 (78 years) was eight yearslonger than that of men2; significantly higher mortality overall was reportedamong tea, rubber and coconut plantation workers (although some decline hasbeen observed more recently)59; and there is higher maternal mortality in thenorthern and eastern districts affected by conflict. For example, in 1995–1996,the maternal mortality ratio in these districts was 3.5 times higher than that ofthe entire country, most likely due to poorer access to health services, education,nutrition, water and sanitation60.



Table 8.5 Selected development indicators, Sri Lanka vs. India

Indicator Sri Lanka Indiaa

1985 2008 1985 2008


70 97b 18 66


83.0 98.4 n/a 93.6c

Poverty gap at $2 a day (purchasing power parity) (%)

16.1 11.9 36.7d 30.4e


86.8f 90.6 40.8f 62.8g

Undenutrition prevalence, weight forage (% of children under 5 years)

29.3h 21.1i n/a 43.5g


2.92 2.33 4.32 2.74


69.0 74.1 56.9 63.7


Notes: n/a: Not available; DPT: Three doses, diphtheria, pertussis and tetanus; aSelected as suitablecomparator due to similar location and World Bank income group; bValue is for 2009; cValue isfor 2007; dValue is for 2002; eValue is for 2005; fValue is for 1981; gValue is for 2006; hValue isfor 1987; iValue is for 2007.

The health system and its performance since 1985

As in other social sectors, the foundation for Sri Lanka’s current health systemwas laid prior to 1960 and no major structural changes have occurred since then(Figure 8.7). Today, the overall health system is composed of parallel public andprivate sectors. The comprehensive public system is financed and operated bythe Ministry of Health in Colombo and eight provincial departments of health,and almost all care from preventive services to specialist tertiary care is free at thepoint of delivery. Units run by medical officers provide most preventive andpublic health services through teams of health workers59. Having grown steadilysince the 1960s, the private sector is also very prominent and focuses mainly onoutpatient care, but there is also a small private hospital sector concentrated inthe capital59. Much private sector activity is actually provided by governmentmedical officers working during their off-duty hours. This practice allows thesepublic servants to supplement their relatively meagre government wages andpromotes the retention of health professionals in the public service. As such, theoverall outpatient load is shared between the public and private sectors, while thepublic sector provides more than 95% of inpatient care59.



0

10

20

30

40

50

60

70

80

90

100

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

1978

1976

1974

1972

1970



Figure 8.6 Infant and maternal mortality, Sri Lanka, 1970–2009


Good access to health services is likely to explain at least some of the country’soverall health performance. After decades of government investment in hospitalinfrastructure, most Sri Lankans live within 3 km of a public facility59, and since2000, there has been an average of three hospital beds per 1000 people(compared with the average of two beds per 1000 people in middle incomecountries)2. Another legacy of this early investment is the low overall spending



Figure 8.7 Timeline of key events influencing health, Sri Lanka, 1960–2010

Private health sector begins togrow

Civil war begins

Development of Health SectorMaster Plan 2004–2015

End of civil war

Economic downturn: food andenergy shortages

Family planning programme operationalized with establish-ment of Family Health Bureau

Period of economic liberalization,scaling back of food programmes

Decentralization of health services from central to provincial ministries of health

Tsunami

1960

1970

1980

1990

2000

2010

on health: in 2007, total health expenditure was estimated to be 4.2% of GDP,slightly lower than other countries at the same income level2. However, just 47%of this expenditure is now public (constituting 8.5% of total government expen-diture), leaving more than half of health spending financed privately, 86% ofwhich comes out of pocket, as private health insurance coverage is low2.Increasing private expenditure is the main driver of increasing health expendi-ture overall. The majority of public funding goes towards the provision of in-patient services, while the bulk of private funding is spent on outpatientcare59. These public–private divisions in provision and spending have impartedto Sri Lanka’s overall health system several interesting performance characteris-tics, most notably with respect to equity and quality.

Good geographic access and lack of financial barriers to public facilities combinewith several aspects of health financing to produce a relatively equitable healthsystem in Sri Lanka. First, public health spending is relatively progressive,reflecting the role of direct taxation. The ability of the rich (who would typicallycontribute more but use services less) to opt out of the public system has tendedto leave public outpatient care dominated by the poor61, resulting in the poorestquintile benefiting from 27% of public spending in this sector, compared with11% for the richest quintile in 2003–200459. However, because of the tendencyto use the public system for inpatient care across all income groups, governmentspending for this type of care is more evenly distributed (18% for the poorestquintile versus 16% for the richest, although this takes no account of the muchgreater health needs of the poor)59.

The high use of public sector inpatient care by all groups provides some protec-tion from catastrophic health expenses by limiting out-of-pocket payments. Onestudy that looked at the incidence of such expenses across a number of Asiancountries showed that only a very small percentage of Sri Lankan householdswere affected, which was much better than in many other low- and middle-income countries62. High utilization of inpatient services in the public sector ismaintained by a perception of good quality based on the widely held view thatpublic hospitals have the best staff and equipment to deal with serious condi-tions61,63. In outpatient clinics, on the other hand, perceived low quality ofpublic sector services persuades richer patients to pay for private care59.

It has been argued that the high standard of training received by practitioners inthe public sector has helped to create and maintain quality in the private sector,since, as a consequence of dual practice, most doctors are drawn from the publicsector and are believed to apply the same practice standards in both sectors. Onestudy conducted in 2001 estimated that between 50% and 70% of the privatesector caseload was being seen by doctors regularly employed in the public



sector64. This means that although use of the private sector has grown it has notinvolved the expansion of unqualified, informal providers, which characterizethis sector in many other developing countries60. Furthermore, because mostprivate practitioners are drawn from public cadres working within the same area,both patients and providers already know which private providers offer a highstandard of care (Russell S, personal communication, 2010). There are, ofcourse, well-known disadvantages arising from dual practice, including theincentive to exert more effort in the more lucrative private sector or even to stimulate private demand by underperforming in public practice65.

Overall, Sri Lankans place a great deal of confidence in their health system –both public and private sectors – because of the quality care, ease of access andlevel of risk protection it provides. Nevertheless, decreasing levels of governmenthealth investment have affected hospital care, which could drive greater numbersof patients to seek care in the private sector. For example, between 1987 and2004, the percentage of patients who sought modern care from the private sector(including clinics, hospitals and pharmacies) increased by nearly 8%, to reach45.1%, while recent analysis of the impact of government health spending indi-cates a shifting of benefit from the poor to the urban better-off59. It is recognizedthat the continued shift of patients out of the public system may destabilize thehealth system by undermining popular support for government health serv-ices59. With the recent end of the longstanding civil war, an opportunity hasarisen for the 12–14% of total government expenditure previously spent on themilitary to be redistributed across other national priorities2.

■ Concluding remarks: Further insights after 25 years?

The original Good health at low cost in 1985 concluded with three recommen-dations for other developing countries seeking to improve population health.Countries should work first to ensure equitable access to public health servicesand health care; second, to provide universally accessible education; and third, toguarantee adequate nutrition to all levels of society. Taken together with otherimportant features identified from the case studies, including universal franchise,promotion of social and economic equality, and development of public infra-structure, these are all products of what has been defined as the ‘support-ledsecurity’ approach to development66. A sustained long-term commitment to thisapproach by each of the governments studied aided the formation of a virtuouscycle of human development that served to build popular trust and confidencein the state’s ability to provide for its citizens’ needs, leading governments to bemore responsive and accountable to these needs.



From examining how China, Costa Rica, Kerala and Sri Lanka have fared 25years later, it is evident that the benefits of these virtuous cycles continue tocompound and have contributed to the impressive health gains observed since1985. While the original lessons clearly continue to be relevant today, revisitingthese case studies in light of the many contextual changes that have occurredsince 1985 has generated new lessons on how health systems can respond tomeet the challenges that are posed not only by increasing burdens of chronicdisease and ageing populations but also by changing economic and social reali-ties, such as migration, values and the ever increasing number of actors involvedin health.

Echoing the first recommendation from the original Good health at low cost,ensuring access to modern health services remains crucial to improving popula-tion health. But in contrast to the early stages of health system developmentwhere the aim was to provide a basic level of care to the entire population, thispriority has broadened to encompass higher levels of care and preventive services inorder to adapt to changing health needs. One of the guiding principles of thisexpansion is guaranteeing that it happens in an equitable way to counterbalancethe effects of changing economic conditions and demographic trends thatcontinually work to widen health inequalities.

Second, achieving equity of access was also shown to be dependent on theacceptability of the care being provided – a concept that has evolved beyond thecultural acceptability as defined by Alma-Ata in 1978 to also include valuesdriven by increasing consumer awareness. In several of the case studies, the rolethat differences in real and perceived quality played was clear in shifting utilization away from readily available care towards more expensive care in eitherthe private sector or in urban areas. Quality was also seen to be important inmaintaining popular confidence in publicly funded health services and publicinstitutions. This confidence is a key factor in keeping health as a political prior-ity and ensuring the financial sustainability of the health system.

Both of these lessons underscore the third emerging theme from the case studies:that governments continue to play a central role in developing health systems. It isinteresting to note that the motivation for the original Good health at low costvolume was partially a reaction to the privatization discourse that dominated thepolitics of the 1980s, and that updating these case studies has once again high-lighted the importance of continued government leadership in developing accessible and responsive health systems. While there is no one-size-fits-allmodel for such participation, the case studies have clearly shown that govern-ments have a variety of tools at their disposal to help to build and maintain equitable access and quality care. Direct provision of health services through



public-sector facilities was an important avenue to ensure equitable access ineach of the case studies. The case studies also described several mechanisms bywhich government involvement influenced the quality of care (e.g. direct invest-ment, regulation, efficiency interventions). But with tightening budgets and theunavoidable increases in private-sector participation and out-of-pocket expendi-tures, perhaps where government involvement can have the greatest impact isthrough the design, implementation and regulation of financial protectionmechanisms that are appropriate for the context, sustainable and pro-poor.

ACKNOWLEDGEMENTS

I would like to thank Gerald Bloom (Institute of Development Studies,University of Sussex), Arnab Acharya (London School of Hygiene & TropicalMedicine), Steven Russell (School of International Development, University ofEast Anglia), Luis Rosero-Bixby (Centro Centroamericano de Población,Universidad de Costa Rica), and Kent Ranson (World Health Organization) fortheir contribution to this work. I would also like thank Dina Balabanova, GillWalt, Anne Mills and Martin McKee, all from the London School of Hygiene& Tropical Medicine, and Joanne McManus for their guidance and helpfulsuggestions.

REFERENCES

1. Halstead SB, Walsh JL, Warren KS, eds. Good health at low cost. New York:Rockefeller Foundation; 1985.

2. World Bank. World development indicators [online]. 2009 (http://ddp-ext.worldbank.org/ext/DDPQQ/member.do?method=getMembers&userid=1&queryId=,accessed 30 May 2010).

3. Hogan MC et al. Maternal mortality for 181 countries, 1980–2008: a systematicanalysis of progress towards Millennium Development Goal 5. Lancet 2010;375(9726):1609–23.

4. Wagstaff A et al. Reforming China’s rural health system. Washington, DC: WorldBank; 2009.

5. Rajaratnam JK et al. Neonatal, postneonatal, childhood, and under-5 mortalityfor 187 countries, 1970–2010: a systematic analysis of progress towardsMillennium Development Goal 4. Lancet 2010; 375(9730):1988–2008.



6. IHME. Datasets. Seattle, WA: Institute for Health Metrics & Evaluation,Department of Global Health, University of Washington; 2010(http://www.healthmetricsandevaluation.org/resources/datasets.html, accessed 21May 2010).

7. Banister J, Zhang X. China, economic development and mortality decline. WorldDevelopment 2005; 33(1):21–41.

8. Audibert M, Mathonnat J, Chen Y. Does external openness influence the infantmortality rates? An econometric investigation for the Chinese provinces. Clermont-Ferrand: Centre d’Etudes et de Recherches sur le Développement International;1999 (CERDI Working Paper: 1998311998).

9. Wagstaff A et al. China’s health system and its reform: a review of recent studies.Health Economics 2009; 18(Special Issue):S7–23.

10. Tang S et al. Tackling the challenges to health equity in China. Lancet 2008;372(9648):1493–501.

11. Sidel VW. The barefoot doctors of the People’s Republic of China. New EnglandJournal of Medicine 1972; 286(24):1292–300.

12. Blumenthal D, Hsiao W. Privatization and its discontents: the evolving Chinesehealth care system. New England Journal of Medicine 2005; 353(11):1165–70.

13. West LA, Wong CPW. Fiscal decentralization and growing regional disparities inrural China: some evidence in the provision of social services. Oxford Review ofEconomic Policy 1995; 11(4):70–84.

14. Grigoriou C, Guillaumont P. Child mortality under Chinese reforms. Clermont-Ferrand: Centre d’Etudes et de Recherches sur le Développement International;2004 (CERDI Working Paper: 2004102004).

15. Sun X et al. Catastrophic medical payment and financial protection in ruralChina: evidence from the New Cooperative Medical Scheme in ShandongProvince. Health Economics 2009; 18(1):103–19.

16. Qingyue M, Shenglan T. Universal coverage of health care in China: challenges andopportunities. Geneva: World Health Organization; 2010.

17. Unger JP et al. Costa Rica: achievements of a heterodox health policy. AmericanJournal of Public Health 2008; 98(4):636–43.

18. UNDP. Human development reports [online]. New York: United NationsDevelopment Programme; 2010 (http://hdr.undp.org/en/, accessed 4 January2011).



19. Cercone J, Pacheco Jimenez J. Costa Rica: “good practice” in expanding healthcare coverage – lessons from reforms in low- and middle-income countries. In:Gottret P, Schieber GJ, Waters HR, eds. Good practices in health financing: lessonsfrom reforms in low- and middle-income countries. Washington, DC: World Bank;2008:183–226.

20. Rosero-Bixby L. [Assessing the impact of health sector reform in Costa Ricathrough a quasi-experimental study.] [In Spanish] Revista Panamericana de SaludPública 2004; 15(2):94–103.

21. McGuire JW. Politics, policy, and mortality decline in Costa Rica. Middletown, CT:Wesleyan University; 2007.

22. deBertodano I. The Costa Rican health system: low cost, high value. Bulletin ofthe World Health Organization 2003; 81(8):626–27.

23. Clark MA. Reinforcing a public system: health sector reform in Costa Rica. In:Kaufman RR, Nelson JM, eds. Crucial needs, weak incentives: social sector reform,democratization, and globalization in Latin America. Baltimore, MD: JohnsHopkins University Press; 2004:189–216.

24. Kruk ME et al. The contribution of primary care to health and health systems inlow- and middle-income countries: a critical review of major primary care initia-tives. Social Science & Medicine 2010; 70(6):904–11.

25. Rosero-Bixby L. Infant mortality decline in Costa Rica. In: Halstead SB, WalshJL, Warren KS, eds. Good health at low cost. New York: Rockefeller Foundation;1985:125–38.

26 Rosero-Bixby L. Socioeconomic development, health interventions and mortalitydecline in Costa Rica. Scandinavian Journal of Social Medicine. Supplement 1991;46:33–42.

27. Homedes N, Ugalde A. Why neoliberal health reforms have failed in LatinAmerica. Health Policy 2005; 71(1):83–96.

28. Costa Rica. Decentralization of hospitals and clinics (GLIN ID 87021). LaGaceta, Diario Oficial, December 1998

29. Centre for Development Studies. Human Development Report 2005.Thiruvananthapuram: State Planning Board, Government of Kerala; 2006.

30. Pushpangadan K. Remittances, consumption and economic growth in Kerala:1980–2000. Trivendrum: Centre for Development Studies; 2003.

31. Narayanamoorthy A, Kamble BN. Trends and determinants of rural literacyamong scheduled caste population: a state level analysis. Journal of EducationalPlanning and Administration 2003; 17(1):35–52.



32. International Institute for Population Sciences. National family health survey(MCH and family planning), 1992–93: India. Bombay: International Institute forPopulation Sciences; 1995.

33. International Institute for Population Sciences and Macro International. Nationalfamily health survey, 2005–06: India, Volume I. Mumbai: International Institutefor Population Sciences; 2007.

34. International Institute for Population Sciences and Macro International. Nationalfamily health survey, India, 2005–06: Kerala. Mumbai: International Institute forPopulation Sciences; 2008.

35. Krishnan T. Health statistics in Kerala State, India. In: Halstead SB, Walsh JL,Warren KS, eds. Good health at low cost. New York: Rockefeller Foundation;1985:39–46.

36. Directorate of Economics and Statistics, Government of Kerala. Annual reports onsample registration. Thiruvananthapuram: Government of Kerala; 1981.

37. International Institute for Population Sciences and ORC Macro. National familyhealth survey (NFHS-2), 1998–99: India. Mumbai: International Institute forPopulation Sciences; 2000.

38. National Rural Health Mission. Kerala State file [online]. Mumbai: Ministry ofHealth & Family Welfare, Government of India; 2010 (http://www.mohfw.nic.in/NRHM/State%20Files/kerala.htm, accessed 1 June 2010).

39. Kumar BG. Low mortality and high morbidity in Kerala reconsidered. Populationand Development Review 1993; 19(1):103–21.

40. Narayana D. High health achievements and good access to health care at greatcost: the emerging Kerala situation. In: Haddad S, Barı E, Narayana D, eds.Safeguarding the health sector in times of macroeconomic instability: policy lessons forlow- and middle-income countries. Trenton: Africa World Press/IDRC; 2008.

41. Soman CR et al. All-cause mortality and cardiovascular mortality in Kerala Stateof India: results from a 5-year follow-up of 161 942 rural community dwellingadults. Asia-Pacific Journal of Public Health 2010; May 10 [Epub ahead of print].

42. WHO. Global status report on alcohol 2004. Geneva: World Health Organization,Department of Mental Health and Substance Abuse; 2004.

43. Panikar P. Health transition in Kerala. Thiruvananthapuram: Kerala ResearchProgramme on Local Level Development, Centre for Development Studies; 1999.

44. Sato H. Social security and well-being in a low-income economy: an appraisal ofthe Kerala experience. Developing Economies 2004; 42(2):288–304.



45. Kutty VR. Historical analysis of the development of health care facilities in KeralaState, India. Health Policy and Planning 2000; 15(1):103–9.

46. Levesque JF et al. Outpatient care utilization in urban Kerala, India. Health Policyand Planning 2006; 21(4):289–301.

47. Nabae K. The health care system in Kerala: its past accomplishments and newchallenges. Journal of the National Institute of Public Health 2003; 52(2):140–5.

48. Dilip TR. Utilization of inpatient care from private hospitals: trends emergingfrom Kerala, India. Health Policy and Planning 2010; 25(5):437–46.

49. Rao KS et al. Financing of health in India (Section IV). New Delhi: NationalCommission on Macroeconomics and Health, Ministry of Health & FamilyWelfare, Government of India; 2005.

50. Kunhikannan T, Aravindan K. Changes in the health status of Kerala 1987–1997.Thiruvananthapuram: Kerala Research Programme on Local Level Development,Centre for Development Studies; 2000.

51. Ashokan A, Ibrahim P. Inpatient health care expenditure: some new evidencesfrom rural Kerala. Indian Journal of Economics and Business 2008; 7(2):297–307.

52. Nair VM. Health in South Asia: future of Kerala depends on its willingness tolearn from past. British Medical Journal 2004; 328(7454):1497.

53. Constitutional (73rd Amendment) Act (1992).

54. Kerala Panchayt Raj Act (13/1994) (1994).

55. Drèze J, Sen A. India: economic development and social opportunity. Oxford:Clarendon Press; 1995.

56. Varatharajan D, Thankappan R, Jayapalan S. Assessing the performance ofprimary health centres under decentralized government in Kerala, India. HealthPolicy and Planning 2004; 19(1):41–51.

57. US Department of State. Background note: Sri Lanka. Washington, DC:Department of State; 2010 (http://www.state.gov/r/pa/ei/bgn/5249.htm,accessed 12 July 2010).

58. Amnesty International. Sri Lanka’s displaced face uncertain future as governmentbegins to unlock the camps. Updated 11 September 2009 (http://www.amnesty.org/en/news-and-updates/news/sri-lanka-displaced-uncertain-future-government-unlock-camps-20090911, accessed 7 July 2010).

59. Rannan-Eliya RP, Sikurajapathy L. Sri Lanka: “good practice” in expanding healthcare coverage. In: Gottret P, Schieber GJ, Waters HR, editors. Good practices inhealth financing: lessons from reforms in low- and middle-income countries.Washington, DC: World Bank; 2008:311–54.



60. McNay K, Keith R, Penrose A. Bucking the trend: how Sri Lanka has achieved goodhealth at low cost – challenges and policy lessons for the 21st century. London: Savethe Children; 2004.

61. Russell S, Gilson L. Are health services protecting the livelihoods of the urbanpoor in Sri Lanka? Findings from two low-income areas of Colombo. SocialScience & Medicine 2006; 63(7):1732–44.

62. van Doorslaer E et al. Catastrophic payments for health care in Asia. HealthEconomics 2007; 16(11):1159–84.

63. Russell S. Treatment-seeking behaviour in urban Sri Lanka: trusting the state,trusting private providers. Social Science & Medicine 2005; 61(7):1396–407.

64. Rannan-Eliya R et al. Equity in financing and delivery of health services inBangladesh, Nepal and Sri Lanka. Colombo: Institute of Policy Studies; 2001.

65. Ferrinho P et al. Dual practice in the health sector: review of the evidence. HumanResources for Health 2004; 2(1):14.

66. Drèze J, Sen A. Hunger and public action. Oxford: Clarendon Press; 1989.



good health at low cost revisited - ghlc...benjamin palafox lshtm chapter 8 good health at low cost...

Documents